February 2017 Summaries

In February 2017, Wesley Chun from the G Suite team introduced a new tutorial for the Google Slides API, highlighting its potential for automating presentation tasks through programmatic interactions. Since the API's launch, developers have been able to manipulate text and images within slides, generate presentations from spreadsheet data, and format text using JSON payloads sent to the batchUpdate() method. The latest tutorial focuses on adding text and shapes to slides, emphasizing the importance of creating custom object IDs to optimize API requests and reduce calls. The process involves using insertText requests for text addition, while adding shapes necessitates a more complex JSON structure due to the mathematical calculations required for proper rendering. The tutorial provides practical advice and references additional resources for tasks like text formatting, search-and-replace functions, and integrating data into slides, aiming to assist developers in building applications that streamline the creation of presentations.

TensorFlow Debugger (tfdbg) is a specialized tool designed to facilitate the debugging of machine learning models within Google's open-source TensorFlow library. By addressing the complexity of debugging during the TensorFlow runtime, where standard Python debuggers fall short, tfdbg offers a command-line interface (CLI) that allows users to inspect and interact with dataflow graph nodes, visualize execution histories, and identify issues such as bad numerical values like infinities and NaNs. The tool enhances the debugging experience by enabling stepper mode for detailed exploration of node outputs, and supports conditional breakpoints for efficient problem identification. Compared to alternative methods, tfdbg requires fewer code changes and provides more comprehensive graph coverage, accelerating model development and debugging workflows. It includes features for offline debugging and integrates with tf.contrib.learn, with a minimum TensorFlow version requirement of 0.12.1. Users can find documentation, report bugs on GitHub, and seek help on StackOverflow, with the project benefiting from contributions by Google's TensorFlow Core/API and Applied Machine Intelligence Teams.

TensorFlow 1.0, announced during the first annual TensorFlow Developer Summit, marks a significant milestone with enhancements in speed, flexibility, and production-readiness. The update introduces the high-level API modules tf.layers, tf.metrics, and tf.losses, as well as the tf.keras module for Keras compatibility, ensuring greater ease in building neural networks. Performance improvements are notable, including a 7.3x speedup for Inception v3 on 8 GPUs and a 58x speedup for distributed training on 64 GPUs, bolstered by the experimental release of XLA, a domain-specific compiler targeting CPUs and GPUs. The release also promises Python API stability, facilitating seamless integration of new features, and offers installation improvements with PyPI compliant pip packages. Additionally, the introduction of the TensorFlow Debugger and new Android demos expands its usability, while the community continues to innovate with tools like Fold for dynamic batching and the Embedding Projector. The announcement highlights the thriving TensorFlow community and its contributions to advancing deep learning technologies globally.

The G Suite Developer Sessions at Google Cloud Next 2017 offer a wide array of opportunities for developers to explore the integration and enhancement of G Suite applications. This event, held in San Francisco, highlights the role of all three pillars of cloud computing—Infrastructure as a Service (IaaS), Platform as a Service (PaaS), and Software as a Service (SaaS). Developers will have access to sessions on automating internal processes with Apps Script and APIs for Google Docs editors, creating custom add-ons for G Suite using Apps Script, and leveraging Google Drive APIs for organizational use. The event also introduces Google App Maker, a low-code development tool for building custom applications. Participants can engage with Google engineers, product managers, and fellow developers to exchange ideas and insights on maximizing the potential of G Suite and Google Cloud Platform.

Google Developers India has launched a YouTube channel specifically designed for Indian developers, featuring original content such as technical tutorials, developer spotlights, and complete Android courses. The channel aims to support India's growing developer community, which is expected to become the largest in the world by 2018, by providing local developers with interviews, inspirational stories, and updates on Google news and APIs. This initiative is part of Google's effort to train 2 million Indian developers and capitalize on the country's burgeoning mobile development scene. The channel includes shows like DevShow India, which updates developers on the latest industry trends, and Skilled to Scaled, a series that follows the journey of app development from inception to monetization.

Google has introduced a feature in its AMP integration in Google Search that allows users to access and share the canonical URL of an AMP document, addressing concerns about URL confusion and content attribution. AMP documents typically have three types of URLs: the original URL, the AMP Cache URL, and the Google AMP Viewer URL, each serving different purposes such as caching and pre-rendering to enhance speed and performance. The AMP Cache URL is used to efficiently deliver content through Google's infrastructure, while the AMP Viewer URL facilitates pre-rendering for an instant loading experience by using a hidden iframe. To mitigate confusion and potential security risks, AMP viewers display a header bar that clarifies the content's original source. Google is committed to improving the AMP experience, including implementing features that make it easier for users to share the original URL and exploring new web platform APIs like the Web Share API to further enhance functionality.

The Associate Android Developer Certification, introduced by Google in 2016, aims to establish a professional benchmark for Android developers worldwide, signifying their competency in the field. The program, announced at Google I/O 2016, has since certified over 322 developers across 61 countries, recognizing their skills through a performance-based exam and an exit interview, which test their ability to implement and debug app features using Android Studio. The certification, priced at $149 USD or 6500 INR in India, allows candidates three attempts to pass the exam within a 48-hour window, after which they partake in an exit interview to further demonstrate their knowledge. Successful candidates earn the title of Google Certified Associate Android Developer, a credential they can highlight on their resumes and social media profiles, gaining access to exclusive community benefits and increased visibility in the job market.